Control Chart of T² Hotelling on Quality Control Activities of Crude Palm Oil (CPO) at PT Cipta Graha Garwita, Seluma Regency, Bengkulu Province

Abstract

PT Cipta Graha Garwita (CGG) is a palm oil producer focused on product quality, especially crude palm oil (CPO) for both food and non-food applications. Despite CGG's good reputation, variability in quality characteristics such as Free Fatty Acid (FFA) and moisture can affect the final quality of CPO. This study aims to apply a statistical quality control system to monitor and improve the consistency of CPO quality using T² Hotelling control charts. Statistical quality control methods ensure that products meet standards by reducing variability. One such tool is the T² Hotelling control chart, effective for monitoring multivariate variables using mean vectors and variance-covariance matrices. This study involves steps from data collection, testing multivariate normality assumptions, calculating T² Hotelling control charts, to determining control limits. Testing for multivariate normality assumptions showed the data met normal distribution criteria. The first and second stage T² Hotelling control charts identified several out-of-control observations. These out-of-control observations were excluded, and further analysis showed that after their removal, all data were within statistical control limits. This study recommends further analysis to determine the causes of out-of-control observations using Ishikawa diagrams and process capability evaluation to ensure consistent product quality.